Daegeon Kim

An artificial intelligence (AI) method based on image deep learning is proposed to predict the entire displacement shape of a structure using the feature of partial displacements. The performance of the method was investigated through a structural test of a steel frame. An image-to-image regression (I2IR) training method was developed based on the U-Net layer for image recognition. In the I2IR method, the U-Net is modified to generate images of entire displacement shapes when images of partial displacement shapes of structures are input to the AI network. Furthermore, the training of displacements combined with the location feature was developed so that nodal displacement values with corresponding nodal coordinates could be used in AI training. The proposed training methods can consider correlations between nodal displacements in 3D space, and the accuracy of displacement predictions is improved compared with artificial neural network training methods. Displacements of the steel frame were predicted during the structural tests using the proposed methods and compared with 3D scanning data of displacement shapes. The results show that the proposed AI prediction properly follows the measured displacements using 3D scanning.

The research was initiated to investigate the performance of fiber‐reinforced concrete for protecting people or assets in the building against the explosion or debris missiles. The fiber‐reinforced concrete has the difficulty with being applied in the actual construction conditions with the normal ready‐mixed concrete system. The fibers for the protection performance require high toughness to endure the huge energy from an explosion, but the large amount of the fiber is required. The required amount of fibers can result in decreased workability and insufficient dispersion of fibers. It has been difficult to apply fiber‐reinforced concrete on field placing with the ready‐mixed concrete system of plant mixing, delivering, and placing. This research carried out the investigation of properties of combined fiber of steel and polymeric fiber to improve workability and agitating in the mixer. Based on the preliminary experimental test results in a laboratory, combined fiber‐reinforced concrete was applied on the actual field construction of chemical plant. According to the results from the laboratory tests and application in the real construction project, it is expected to introduce the combined fiber for desirable mechanical performance with less adverse effect on workability of the mixture.

The massive construction projects such as nuclear power plants and gas storage plants operate their own batch plants for appropriate supply of a massive amount of ASTM Type I cement. The batch plants have difficulty responding to diversifying consumer requirements such as early strength development of concrete under the current limited production line of cement. In order to respond to the needs, this study collected several sizes of cement particles from different filters at a cement plant, and conducted the adjustment of cement particle‐size distribution to enhance the properties of concrete based on ASTM Type I cement. This paper shows the chemical properties and physical tests such as setting time and compression tests considering the effects of the distribution of cement particles. Also, the classification based on cement particle‐size distribution was defined as FMC (fineness modulus of cement). As FMC increased, compressive strength was relatively low at early age but the difference became smaller for later age. The test results show the effects of the adjusted mix proportion considering particle‐size distribution of cement on strength development of concrete. Therefore, it is possible to see that customized cement or concrete can be manufactured in response to various consumer requirements by introducing such method that by‐passes and collects fine or coarse cement in the cement‐crushing process and then re‐mixes them with OPC ASTM Type I cement depending on FMC.